Risk Measurement In Electricity Markets Based On The Heterogeneous Market Hypothesis

With the electricity market reform in recent decades, we have witnessed the increasing level of electricity price volatility with greater uncertainty and risks. Traditional risk measurement methodologies are constructed under the Efficient Market Hypothesis (EMH).But recent results from empirical studies have provided the evidence of the existence of extreme events and heterogeneous market structures, where the Heterogeneous Market Hypothesis (HMH) fit the market better than the traditional EMH. In this framework, this paper introduces the Empirical Mode Decomposition (EMD) model to study the risk movement in the electricity market and estimate Value at Risk (VaR). EMD model decomposes the time series into several Intrinsic Mode Functions (IMFs), which are categorized based on their different characteristics. Then they are individually modeled by the traditional EWMA model to estimate single asset VaR. Moreover, the Bivariate EMD model is also introduced to estimate Portfolio VaR (PVaR) and analyze the multi-scale and the volatility behaviors of the correlation between two different electricity markets. The decomposed time series will be calculated with the multivariate GARCH model. Empirical studies in the Australian electricity markets suggest that both of the proposed models based on EMD model and BEMD modelrespectively outperform the benchmark EWMA and GARCH model, in terms of conventional performance evaluation criteria for the model reliability.